sas矩阵计算 - 经管之家

5247

收藏 2012-04-01

我用sas计算出一个行列式的值和matlab和excel计算的值不一样，各位大侠帮解释一下
用sas计算的结果是0，用execel和matlab计算的结果是-0.1003（大概是这个数，肯定不等于0）
直接用det(cov)计算的结果是0，
我自己写了一个计算行列式的程序计算的结果也是0.程序如下
do i=1 to 6;
do j=i+1 to 6;
  cov[,j]=cov[,j]-cov[i,j]/cov[i,i]*cov[,i];
  end;
  end;
det=cov[1,1]*cov[2,2]*cov[3,3]*cov[4,4]*cov[5,5]*cov[6,6];
  print cov det;
quit;
下面是cov矩阵
                                       cov
                              COL1       COL2       COL3       COL4       COL5       COL6
                  ROW1 455.06326 -62.93587 435.78672 78.720136 445.33406 -87.36809
                  ROW2 -62.93587 39.397144 -10.07726 46.266692 -38.98252 26.293196
                  ROW3 435.78672 -10.07726 532.52756 181.45362 407.05368 -96.68608
                  ROW4 78.720136 46.266692 181.45362    168.5219 113.79636 59.125268
                  ROW5 445.33406 -38.98252 407.05368 113.79636 554.41088    9.41544
                  ROW6 -87.36809 26.293196 -96.68608 59.125268    9.41544 149.84354

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

全部回复

bobguy

2012-4-2 02:53:46

The sas/iml returns 0.0000507 for  the matrix determinant of Cov.  It is the same as that fro MS.

*****MS syntax
Syntax

MDETERM(array)

Array is a numeric array with an equal number of rows and columns.

**********SAS/IML
  detCov

  0.0000507

proc iml;
cov={
455.06326 -62.93587 435.78672 78.720136 445.33406 -87.36809,
-62.93587 39.397144 -10.07726 46.266692 -38.98252 26.293196,
435.78672 -10.07726 532.52756 181.45362 407.05368 -96.68608,
78.720136 46.266692 181.45362    168.5219 113.79636 59.125268,
445.33406 -38.98252 407.05368 113.79636 554.41088    9.41544,
-87.36809 26.293196 -96.68608 59.125268    9.41544 149.84354
};

detCov=det(cov);

print detCov;
quit;